A Proton Exchange Membrane ("PEM") fuel cell converts the chemical energy of fuels such as hydrogen and oxidizers such as air/oxygen directly into electrical energy. The PEM is a solid polymer electrolyte that permits the passage of protons (H.sup.+ ions) from the "anode" side of a fuel cell to the "cathode" side of the fuel cell while preventing passage therethrough of the hydrogen and air/oxygen gases. Some artisans consider the acronym "PEM" to represent "Polymer Electrolyte Membrane." The direction, from anode to cathode, of flow of protons serves as the basis for labeling an "anode" side and a "cathode" side of every layer in the fuel cell, and in the fuel cell assembly or stack.
For instance, the PEM can be made using a polymer such as the material manufactured by E. I. Du Pont De Nemours Company and sold under the trademark NAFION.RTM.. Further, an active electrolyte such as sulfonic acid groups is included in this polymer. Also, the PEM is available as a product manufactured by W.L. Gore & Associates (Elkton, Md.) and sold under the trademark GORE-SELECT.RTM.. Moreover, a catalyst such as platinum which facilitates chemical reactions is applied to each side of the PEM. This unit is commonly referred to as a membrane electrode assembly ("MEA"). The MEA is available as a product manufactured by W.L. Gore & Associates and sold under the trade designation PRIMEA 5510-HS.
An individual fuel cell generally has multiple, transversely extending layers assembled in a longitudinal direction. In the fuel cell assembly or stack, all layers that extend to the periphery of the fuel cells have holes therethrough for alignment and formation of fluid manifolds. Further, gaskets seal these holes and cooperate with the longitudinal extents of the Layers for completion of the fluid manifolds. As is well-known in the art, some of the fluid manifolds distribute fuel (e.g., hydrogen) and oxidizer (e.g., air/oxygen) to, and remove unused fuel and oxidizer as well as product water from, fluid flow plates of each fuel cell. Furthermore, other fluid manifolds circulate water for cooling.
As is well-known in the art, the PEM can work more effectively if it is wet. Conversely, once any area of the PEM dries out, the fuel cell does not generate any product water in that area because the electrochemical reaction there stops. Undesirably, this drying out can progressively march across the PEM until the fuel cell fails completely.
Traditionally, attempts have been made to introduce water into the PEM by raising the humidity of the incoming reactant gases. That is, the fuel and oxidizer gases are often humidified with water vapor before entering the fluid manifolds in order to convey water vapor for humidification of the PEM of the fuel cell.
However, problems can result from the use of water vapor in humidification of the reactant gases. For example, significant quantities of heat are required in order to saturate the reactant gas stream at a temperature close to the temperature of the fuel cell. In particular, one cannot just employ waste heat from a cell cooling circuit, because the temperature will necessarily be lower than the cell temperature. Furthermore, temperature variations within the reactant gas manifolds and fuel cell plate channels can undesirably lead to condensation of the vapor and poor distribution of the reactant gas and vapor/water.
Moreover, vapor distribution is unpredictable. So, despite the introduction of water vapor into the gas stream at the inlet of a longitudinal fluid manifold, drying out of the PEMs can still occur. These drying problems of the fuel cell assembly can become severe at high power levels.
Deleterious effects can also result from turns in the flow path of a stream which is a mixture of water droplets and reactant gas. After the stream goes around a given curve, separation of the water from the reactant gas occurs. Anytime the stream changes direction and/or velocity, the various settling rates yield separation. Therefore, by the time the stream reaches the end of such a flow path, most of the liquid will have settled out. Similar problems and unpredictability can result in any unconstrained flow of water mixed with reactant gas.
Naturally, fuel cells within the same assembly or stack can have varying efficiencies. In particular, some fuel cells generate more heat than others. A fuel cell running hot will require more water in order to function. If a fuel cell assembly delivers inadequate moisture to a given fuel cell, then the PEM of that fuel cell begins to dry out, which causes it to run hotter still since the remaining fuel cells in the assembly continue to force high current therethrough. When the PEM of a fuel cell completely dries out, that fuel cell begins to dry out any adjacent fuel cells. Accordingly, it is desirable to deliver water to all the fuel cells in the stack.
Additional problems stem from height variations in different areas of an individual fuel cell and the fuel cell assembly. For example, one can consider a fuel cell assembly that is angled and sloping upward from its entry end of a longitudinal reactant fluid supply manifold. There, the mere injection of water at the entry of the manifold into the fuel cell assembly would undesirably result in fuel cells on the low end receiving all water and no gas ("PEM flooding"), and fuel cells on the high end receiving all gas and no water ("PEM starvation").
In one prior art attempt to address some of the problems outlined above, a system is designed so various waters cool the fuel cells in a stack and hydrate their respective PEMs. A hydrogen gas stream is delivered to each anode plate. Injection ports from a water line mix liquid water into a given hydrogen gas stream before it arrives at an anode plate. The number of injection ports determines the amount of water injected into the gas stream, which thereafter flows to the anode plate. Nevertheless, there is no guarantee that every flow channel of the anode plate will obtain from the humidified gas stream adequate water for hydration of its part of the PEM. Furthermore, the possibility of uneven delivery of water among flow channels represents a potential waste in the system. Such a design is disclosed in U.S. Pat. No. 4,769,297 to Reiser et al. (entitled "Solid Polymer Electrolyte Fuel Cell Stack Water Management System," issued Sep. 6, 1988, and assigned to International Fuel Cells Corporation).
Thus, a need exists for ensuring effective, efficient, and continuous supply of water to all active areas of a membrane of an individual fuel cell. A further need exists for ensuring all active areas of each membrane in a working section of a fuel cell assembly effectively and efficiently receive adequate water.